Compilation and Assessment of Lab Samples from EGCS … · • In 2016 Carnival Corporation...

Compilation and Assessment of Lab Samples from EGCS Washwater Discharge on

Carnival ships

February 2019

EXECUTIVE SUMMARY

2

3

• In 2016 Carnival Corporation collected an initial set of 79 Exhaust Gas Cleaning System (EGCS)wash-water samples from 23 ships.

• Samples were collected by shipboard staff using a standard US EPA-referenced samplingprotocol and training, and analysed for 54 test parameters at independent SGS environmentalspecialist ISO certified laboratories, all using EPA standard methods.

• DNV GL’s Maritime Advisory Services were engaged by Carnival to compile and review thelaboratory test data and to compare the washwater lab results to various national andinternational point source discharge limits and water quality standards.

• Since the completion of the 2016 study, Carnival has continued to take additional samples - atotal number of 281 by early 2018, taken from 53 ships.

• In 2018, the assessment was updated and re-evaluated to incorporate these additionalstandards.

• The objective of this work is to better to understand the quality of EGCS wash-water and theparameters which are present.

• As no water quality standards contain criteria for all the EGCS test parameters, in bothassessments the results were compared to both “point source” and “water quality” standards inorder to gain the fullest understanding of the potential EGCS impact on water quality.

Background to the work

Sampling and analysis methods

4

Sampling

• Samples are collected and analyzed in accordance with a standard US EPA-referenced Protocoldeveloped in cooperation with SGS. Samples are taken by ship’s Environmental Officers andEngineers following an onboard training program.

• The sample data base includes samples taken at the inlet, EGCS tower, and outlet.

• The expanded sample data base also includes samples from 11 ships with wash-water filtrationinstalled, providing a fourth sample from a point before this filter.

Analysis

• Sample analysis is conducted by ISO 17025 certified SGS laboratories using US EPA approvedmethods (or equivalent) to test for 54 parameters, including PAHs and metals.

• A “net post-EGCS” methodology is used to compare to various water quality standards, with theadditional analytic step of using a trimmed mean excluding statistical outliers more than threestandard deviations from the mean.*

• The laboratory results have been compiled and reviewed by DNV GL. DNV GL were not involvedin the sampling process or the laboratory analysis of samples.

*This is consistent with the United States Geological Survey’s Statistical Methods in Water Resources.

5

Carnival Open Loop EGCS (DeSOx)

NOTE: Sampling points are installed at the seawater inlet (1), the DeSOx outlet (2) and before the overboard discharge (3).Net post-EGCS values were calculated by comparing the DeSOx outlet values (2) with SW inlet values (1).

6

(Phenanthrene)

Total detected PAH compounds

0%

20%

40%

60%

80%

100%

PAH Nitrate

Average gross PAH and nitrate concentrations as % of IMO criteria

• The wash-water discharge

limitations vary according to the

EGCS wash-water output, i.e. the

higher the wash-water flow the

lower the allowed concentration.

• Here, the weighted average gross

post-EGCS concentrations are

compared with the lowest

theoretical requirements.

• Both the Post-EGCS PAH values for

phenanthrene and the average

sum of all detected PAH values

are below the strictest IMO

requirements.

• This comparison is for illustrative

purposes, and does not constitute

approval of any vessel with IMO

requirements.

CompoundLowest IMO criteria

(based on 90 t/MWh flow rate)Average gross post EGCS concentration

PAH 25 µg/L 1.77 µg/L *

Nitrate 30 mg/L 0.8 mg/L

Note: IMO requirements for PAH are based on phenanthrene equivalence, so the post-EGCS value given is

for phenanthrene. Total detected PAH compounds are shown on the graph for information.

Sample analysis shows average PAH and nitrate levels below IMO wash-water criteria

Comparison to water quality standards gives a deeper understanding of wash-water composition

7

• Comparing the wash-water to various water standards does not intend to indicate compliance with those - only to illustrate the quality in a way that is easy to understand.

• Comparison is made to various standards including:• The German Wastewater Ordinance: Point source

wastewater limitations from biological waste treatment. • The EU Surface Water Standards (part of EU Water

Framework Directive): A water quality standard showing the maximum allowed concentration in inland surface waters.

• This method uses the difference between the average of inlet values and the average of post-EGCS values (sampling point 2 – point 1).

• All non-detected sample parameters are given a value equal to 50% of the laboratory equipment detection limit.

• Results shown exclude statistical outliers more than three standard deviations (3σ) from the mean sample parameter value.

0%

20%

40%

60%

80%

100%

EU Surface Water Standard

0%

20%

40%

60%

80%

100%German Wastewater Ordinance

8

• The study presents a snapshot of average wash-water concentrations across a significant fleet of operating EGCS equipped ships.

• The study provides an objective evaluation of wash-water quality as the basis for more informed debate. It does not:• Attempt to assess the cumulative effect of wash-water parameters entering seawater or the potential

environmental impact.

• Make any conclusions regarding wash-water parameter concentrations and fuel types, engine and flow conditions.

• All laboratory analysed samples remain comfortably within the minimum IMO criteria for pAHand nitrate concentrations.

• As with the earlier 2016 study, the concentrations of all tested parameters for the 2018 samples are below the criteria for relevant land-based point source waste water standards.

• In addition to point source standards, the samples are compared to standards with stricter criteria. These standards are not directly applicable to point source discharges but provide a broader context for understanding wash-water quality. When statistical outliers with a three-sigma deviation are excluded through a trimmed mean statistical method, the wash-water parameters are below these criteria.*

Summary

*Only a low proportion of samples are excluded by this method and many of these statistical outliers are present in the inlet as well as post-EGCS samples.

• Evaluating wash-water beyond compliance

• Assessment approach

• Limitations of the assessment

• The 2016 and 2018 assessments

9

• Executive Summary

• Background

• Sampling and Analysis Methodology

• Observations from Lab results

• Comparison to water quality standards

• Conclusions

Evaluating wash-water beyond compliance

10

• Carnival Corporation employs continuous monitoring of required Exhaust Gas Cleaning System (EGCS) wash-water parameters according to:

• IMO requirements (MEPC.259(68) 2015 Guidelines for exhaust cleaning systems)

• EU Sulphur Directive (EU/2012/33)

• US EPA Vessel General Permit rules

• All Carnival vessels are equipped with continuous monitoring equipment to log required parameters automatically.

• The IMO encourages periodic monitoring to test for additional parameters with spot samples, using measurement techniques which are more advanced than those possible to achieve during continuous monitoring.

• In 2016 a total of 79 samples were collected and analysed from the first 23 Carnival EGCS vessels.

• Sampling continued and in 2018 a re-analysis was undertaken on a total of 281 samples taken from 53 ships.

Vessels are encouraged to perform periodic monitoring for:*

‐ pH‐ PAH and oil‐ Nitrate‐ Nitrite‐ Cadmium‐ Copper‐ Nickel‐ Lead‐ Zinc‐ Arsenic‐ Chromium‐ Vanadium

Periodic monitoring is separate from the required continuous monitoring.

* The US VGP also requires periodic monitoring for selenium and thallium.

Assessment Approach

11

• Samples were collected by shipboard staff using a standard US EPA-referenced sampling protocol and training, and analysed for 54 test parameters at independent SGS environmental specialist accredited laboratories, all using EPA standard methods.

• DNV GL’s Maritime Advisory Services were not involved in the sampling process or the laboratory analysis, but were engaged by Carnival to:

• Compile and review the lab test data; and

• Compare wash-water test parameters to various water quality standards.

• Since no water quality standards exist which contain criteria for all parameters tested in the sampling campaign, 2 land-based point source discharge standards were chosen in addition to the IMO’s EGCS wash-water requirements to provide a benchmark for total metal concentrations.

• A comparison was also made to water quality standards as a means of evaluating PAH levels in context. While the water quality standards refer to bodies of water rather than a discharge, they illustrate the general quality of the wash-water discharge and provide a broader perspective than comparison to IMO requirements, or simply a compilation of the average concentration alone can provide.

• Non-weighted averages of the net Post-EGCS concentration for each parameter have been calculated and compared to the relevant criteria limits for these water quality standards.

Limitations of the assessment

12

• Comparing the wash-water to various water standards does not intend to indicate compliance with those - only to illustrate the quality in a way that is easy to understand.

• The study and comparison with water quality standards does not constitute an environmental impact assessment.*

• The objective of this study was to present a snapshot of average wash-water concentrations. It was not within the scope of this assessment to identify relationships or correlations between parameter concentration in the wash-water and the following factors:

• Fuel type and quality

• Fuel consumption

• Dilution rates

• Flow rates of sea water inlet, post EGCS tower and overboard discharge

• Engine loads at sampling

• The test results from the SGS laboratory were taken at face value, assumed to be in accordance with relevant laboratory standards.

*Areas for further study planned by Carnival include: quantification of the accumulation of wash-water parameters entering seawater; determine the potential environment impact on marine life.

The 2016 and 2018 Assessments

13

• The same EPA-referenced standard sampling protocol and training were used in gathering the samples usedfor the 2016 and the 2018 assessments.

• The same “net post-EGCS” methodology was used as in the 2016 DNVGL Assessment, with the additionalanalytic step of using a trimmed mean* excluding statistical outliers more than three standard deviationsfrom the mean.

• In addition to samples at the inlet, EGCS tower, and outlet; the expanded sample data base also includessamples from some ships with wash-water filtration installed, providing a fourth sample from after this filter.

• DNV GL was again engaged by Carnival to review the laboratory analysis results and make the comparison towater quality standards for the 2018 expanded sample base.

• During the 2016 assessments, comparison was made to water quality standard criteria for:

• German waste water ordinance (Article 2 of 6th Ordinance for Amendment of Waste Water Ordinance, Federal Water Act)

• EU surface water standards (Directive 2013/39/EU, amending Water Framework Directive 2000/60/EU)

• EU drinking water standards (Council Directive 98/83/EC)

• US EPA drinking water standards (Clean Water Act, nickel criteria only)

• In the 2018 assessment, EU Industrial Emissions Directive standards for incineration plant waste gas cleaning

waste water were included in the comparison. The EU drinking water standards are in the process of being

revised and updated with reference to the World Health Organisation (WHO) Guidelines for Drinking-water

quality (Fourth Edition) and therefore the 2018 assessment instead makes a comparison to WHO criteria.

*This is consistent with the United States Geological Survey’s Statistical Methods in Water Resources.

• Sampling campaign protocol

• Samples custody and lab analysis

• Process of sampling and analysis results

• Parameters tested for by SGS

• Handling of “Non-Detects”

• Calculation of “net Post-EGCS” concentrations

14


• Background




• Conclusions

Sampling Campaign Protocol Overview

2

Samples are collected and analyzed in accordance with US EPA guidelines• Protocol was developed in cooperation with SGS, which has

a worldwide network of fully accredited laboratories, and is based upon protocols used in published wash-water sample studies and is consistent with the guidelines set forth in US 40 CFR Part 136 and the US EPA Handbook for Sampling and Sample Preservation of Water and Wastewater Under the US Clean Water Act

• Carnival provides an onboard sampling training program for ship’s Environmental Officers and Engineers that includes a Sampling Manual with Sampling Protocol and tailored training video series, to ensure that samples are consistently collected per the approved standard procedures.

• Sample kits are prepared and supplied by SGS, each kit contains pre-labeled sample bottles, a Sample Collection Checklist, Sample Receipt and Offloading Checklist, and Chain of Custody documentation.

• Sample integrity is maintained from collection to analysis following required temperature, handling, and hold times, detailed in the sampling protocol.

15

Samples Custody and Lab Analysis

16

Chain of Custody

• Time/signature Chain of Custody is maintained throughout the sampling, transfer, and laboratory analysis.

Laboratory Analysis• The SGS labs analyze all samples utilizing US EPA

approved methods and issues a final report with all required calibration, standardization, and Limits of Quantitation

• Upon receipt at the laboratory a checklist is created by SGS to document condition of sample kits, including temperature, breakage, custody seals, and confirmation that all samples are accounted for.

• Samples are analyzed within the hold time required by the given method.

• Results provided by the SGS laboratories include the analysis methodology used and the relevant calibration and standardization procedures.

• SGS provides limits of quantitation for each result obtained.

Processing of Sampling and Analysis Results

17

Documentation and process overview

• Final reports are issued by SGS upon completion of sample analysis.

• The reports and sampling process are reviewed by SGS, each operating line, in addition to an independent review.

• Results are evaluated to ensure that sampling procedure was followed, Chain of Custody was maintained, approved analytical methods were utilized, and that hold times were not exceeded.

Compilation of results

• The SGS laboratory reports are submitted to DNV GL to be compiled. DNV GL then reviews the data and makes the comparison to selected water quality standards.

Process adjustments

• To ensure the most standard laboratory treatment possible, Carnival has reduced the laboratories for most samples to two: one in Europe and one In North America.

• Additionally, SGS has cooperatively reduced the detection limits to the lowest level possible for each parameter, with identical limits at both labs.

Parameters tested for by SGS

18

PAHs Metals* Other parameters

1-Methylnaphthalene2-Methylnaphthalene

AcenaphtheneAcenaphthylene

Benzo(a)anthraceneBenzo(a)pyrene

Benzo(b)fluorantheneBenzo(g,h,i)peryleneBenzo(k)fluoranthene

ChryseneDibenz(a,h)anthracene

FluorantheneFluorene

Indeno(1,2,3-cd)pyreneNaphthalene

PhenanthrenePyrene

Aluminum (Al)Cadmium (Cd)Chromium (Cr)

Copper (Cu)Iron (Fe)Lead (Pb)Nickel (Ni)

Thallium (Tl)Vanadium (V)

Zinc (Zn)Arsenic (As)

Selenium (Se)Mercury (Hg)

C10 – C40 HydrocarbonsChloride

Total Dissolved SolidsTotal Suspended Solids

Ammonia as NTotal Phosphate as PO4Total Phosphorus as PTotal Organic Carbon

Total Kjeldahl NitrogenChromium (VI)

pHNitrate + Nitrate as N

Biological oxygen demand (BOD)Chemical oxygen demand (COD)

Organic NitrogenTotal Nitrogen

AmmoniumGasoline Range Organics (GRO)

* Analysis was for both Total and Dissolved portion

Handling of “Non-Detects”

19

• A significant number of samples for each parameterare “non-detects” below the laboratory detectionlimits. In environmental testing, it is technicallyimpossible to verify the complete absence of ananalyte.

• If the analyte is not present or present at a concentration below the detection limit, the lab will report a non detect (ND).

• An analyte is only confidently reported as present when the test result is above the detection limit.

• SGS provided the DL for each analyte and sample

• For results reported as ND, it was assumed that the concentration of the analyte in the sample was half of the detection limit (DL).*

• This means the concentration of the analyte is most likely inflated when calculating the average concentration since a ND can actually be zero, while from a chemical analytical point of view, it cannot be identified definitively as zero.

* US EPA Ref. Statistical analysis of groundwater monitoring data at RCRA facilities: unified guidance (2009)

• Certain parameters are nearly always below the detection limit at the post tower outlet sample.

• An ND is considered to be half the DL, so the result has more uncertainty, although the margin of error rests below the DL

• For a given parameter, the lower the rate of detection rate, the more likely the average concentration is below the DL but the greater the margin of error.

Calculation of “net Post-EGCS” concentrations

20

• In some cases, the net concentration is negative. This negative result is taken at face value, assumed to be validbecause there is also a filter between the seawater inlet and the EGCS tower.

• The intention of using the net concentration is to correct for the amount already present in the incomingseawater, giving a clearer idea of how the EGCS contributes to the change in concentration of the parameter

• If the ambient seawater already has a high concentration of the parameter in question, the concentration afterthe DeSox tower (and in the overboard discharge) may be elevated, but not necessarily due to the EGCSprocess.

• In order to understand how the EGCS process affects water quality*, a distinction is made between gross postEGCS concentration and net Post-EGCS concentration.

Net post-EGCS concentration =

Post-EGCS concentration (2) – Incoming seawater concentration (1)

EGCS DeSOx

towerSeawater inlet

Overboard

discharge

Sample Point

(1)

Dilution process (using sea water)

Sample point

(2)

Sample Point

(3)

* Compliance with IMO wash-water discharge criteria uses the gross post-EGCS concentration.

• Distribution of “non-detects” and statistical outliers.

• Inlet and post-EGCS sample distributions for Anthracene

• Inlet and post-EGCS sample distributions for Nickel

21


• Background




• Conclusions

Observations from the lab results

22

• In both the 2016 and 2018 assessment sample sets, many parameters have a detection rate of less than 50%.

• For certain parameters, the criteria in water quality standards can be below what is possible to measure using known laboratory techniques.

• This does not mean that the results are invalid. The purpose of this assessment is to illustrate the quality of wash-water, not definitively demonstrate compliance with the standards which are compared to.

• The detection limit for a particular sample parameter is dependent upon laboratory analyst, sample matrix, method used, and instrumentation utilized.

• The percentage of samples excluded as being more than three standard deviations from themean varies between 0.4% (1 sample) and 3.2% (9 samples). For the majority of cases(considering all parameters at inlet and post-EGCS sampling points), 1.1% of the samples areexcluded (3 samples).

• The following slides show:

• Distribution of “detects”, “non-detects” and statistical outliers for inlet and post-EGCS samples.

• Inlet and post-EGCS sample distributions for Anthracene and Nickel

Inlet samples: laboratory “detects” vs. “non-detects” Between 60-100% of Inlet samples for each parameter are “non-detects”

23

268 270 267 268 270 268 268 270 266 270 268 268 270

246258 264

160

220

179164

225

250

224

167

212 213

167 164

228215

167

248 249

232

167

214204

179

24 12108 50 88 106 45 46 103 55 57 103 106 42 55 100 22 38 103 53 66 91

3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 6 3 3 3 9 3 3 3 9 3 2 3 3 4 4 5

0

50

100

150

200

250

300

Ace

nap

hth

ene

SW

Inle

t

Ace

nap

hth

ylen

e SW

Inle

t

An

thra

cen

e S

W In

let

Be

nzo

(a)a

nth

race

ne

SW In

let

Be

nzo

(a)p

yre

ne

SW In

let

Be

nzo

(b)f

luo

ran

then

e SW

Inle

t

Be

nzo

(g,h

,i)p

ery

len

e S

W In

let

Be

nzo

(k)f

luo

ran

then

e SW

Inle

t

Ch

ryse

ne

SW In

let

Dib

en

z(a,

h)a

nth

race

ne

SW

Inle

t

Flu

ora

nth

ene

SW In

let

Flu

ore

ne

SW

Inle

t

Ind

eno

(1,2

,3-c

d)p

yren

e SW

Inle

t

Nap

hth

ale

ne

SW

Inle

t

Ph

enan

thre

ne

SW In

let

Pyr

ene

SW

Inle

t

Ars

enic

SW

Inle

t

Cad

miu

m S

W In

let

Ch

rom

ium

SW

Inle

t

Co

pp

er S

W In

let

Lead

SW

Inle

t

Me

rcu

ry S

eaw

ate

r In

let

Nic

kel S

W In

let

Sele

niu

m S

W In

let

Thal

lliu

m S

W In

let

Van

adiu

m S

W In

let

Zin

c SW

Inle

t

Ars

enic

SW

Inle

tD

Cad

miu

m S

W In

letD

Ch

rom

ium

SW

Inle

tD

Co

pp

er S

W In

letD

Lead

SW

Inle

tD

Me

rcu

ry S

eaw

ate

r In

letD

Nic

kel S

W In

letD

Sele

niu

m S

W In

letD

Thal

lliu

m S

W In

letD

Van

adiu

m S

W In

letD

Zin

c SW

Inle

tD

Non-detects Detects # samples excluded

*Statistical outliers > 3σ from the mean

*

Post-EGCS samples: laboratory “detects” vs. “non-detects”Between 6-98% of post-EGCS samples for each parameter are “non-detects”

24

164

224239

156

253

198

240

274

84

264

150

67

263

136

168

228

160147

249259

114

167

225

98

155

178

232

182

139

249256

129

171

226

113

151

116 56 41 124 27 82 38 196 16 130 213 17 264 256 144 111 53 118 134 32 167 114 53 183 126 103 49 99 142 31 152 110 52 168 130

3 4 3 3 3 3 3 3 3 3 3 3 1 2 6 3 3 3 3 3 3 1 2 3 3 2 1 3 3 2 1 4 2 2 4 3 2 1

0

50

100

150

200

250

300

Ace

nap

hth

ene

DeS

ox

Tow

er

Ace

nap

hth

ylen

e D

eSo

x To

we

r

An

thra

cen

e D

eSo

x To

wer

Be

nzo

(a)a

nth

race

ne

De

sox

Tow

er

Be

nzo

(a)p

yre

ne

De

sox

Tow

er

Be

nzo

(b)f

luo

ran

then

e D

eso

x To

we

r

Be

nzo

(g,h

,i)p

ery

len

e D

eso

x To

wer

Be

nzo

(k)f

luo

ran

then

e D

eso

x To

we

r

Ch

ryse

ne

De

sox

Tow

er

Dib

en

z(a,

h)a

nth

race

ne

Des

ox

Tow

er

Flu

ora

nth

ene

De

sox

Tow

er

Flu

ore

ne

Des

ox

Tow

er

Ind

eno

(1,2

,3-c

d)p

yren

e D

eso

x To

wer

Nap

hth

ale

ne

Des

ox

Tow

er

Ph

enan

thre

ne

De

sox

Tow

er

Pyr

ene

Des

ox

Tow

er

Ars

enic

DeS

ox

Tow

er

Cad

miu

m D

eSo

x To

we

r

Ch

rom

ium

De

Sox

Tow

er

Co

pp

er D

eSo

x To

we

r

Lead

De

Sox

Tow

er

Me

rcu

ry D

eSo

x To

we

r

Nic

kel D

eso

x To

wer

Sele

niu

m D

eSo

x To

wer

Thal

lium

De

Sox

Tow

er

Van

adiu

m D

eSo

x To

we

r

Zin

c D

eSo

x To

we

r

Ars

enic

DeS

ox

Tow

erD

Cad

miu

m D

eSo

x To

we

rD

Ch

rom

ium

De

Sox

Tow

erD

Co

pp

er D

eSo

x To

we

rD

Lead

De

Sox

Tow

erD

Me

rcu

ry D

eSo

x To

we

r D

Nic

kel D

eso

x To

wer

D

Sele

niu

m D

eSo

x To

wer

D

Thal

lium

De

Sox

Tow

erD

Van

adiu

m D

eSo

x To

we

rD

Zin

c D

eSo

x To

we

rD

Non-detects Detects # samples excluded

*Statistical outliers > 3σ from the mean

*

25

• Average : 0.2780 μg/l• Average excluding samples outside 3ơ of the mean : 0.1308 μg/l• # samples excluded : 2 out of 281 (0.7%)

Anthracene Inlet Sample Distribution0.7% of the samples are outside 3ơ of the mean

mean + 3ơ

mean - 3ơ

3 samples above laboratory detect limit, including 2 outliers.

26

Anthracene Post-EGCS Sample Distribution1.1% of the samples are outside 3ơ of the mean

• Average : 0.4439 μg/l• Average excluding samples outside 3ơ of the mean : 0.2176 μg/l• # samples excluded : 3 out of 281 (1.1%)

mean + 3ơ

mean - 3ơ


Nickel Inlet Sample Distribution2.1% of the samples are outside 3ơ of the mean

27

• Average : 0.05195 mg/l• Average excluding samples outside 3ơ of the mean : 0.0346 mg/l• # samples excluded : 6 out of 281 (2.1%)

mean + 3ơ

mean - 3ơ

Nic

ke

l In

let

Nickel Inlet

46 samples above laboratory detect limit, including 3 outliers. Remaining 3 outliers are non-detects.

Nickel Post-EGCS Sample DistributionPost-DeSOx samples - 0.7% of the samples are outside 3ơ of the mean

28

• Average : 0.10495 mg/l• Average excluding samples outside 3ơ of the mean : 0.07032 mg/l• # samples excluded : 2 out of 281 (0.7%)

mean + 3ơ

mean - 3ơ

Nic

ke

l p

ost-

De

SO

x

Nickel post-DeSOx


• IMO criteria for EGCS washwater

• Comparison to IMO PAH and Nitrate criteria

• Selection of water quality standards

• Average net post-EGCS analysis compared to:

• German wastewater ordinance

• EU incineration waste water standards

• EU surface water standards

• Positive preliminary results for washwater filtration

29


• Background




• Conclusions

IMO criteria for EGCS wash-water

30

• Nitrates and PAH are regulated according to the output of the EGCS in terms of tons of water per mega watt hour (t/MWh) installed engine power, i.e. the limit is dependent on water output from the EGCS.*

• pH of the discharge point must be the value which will give at least 6.5 pH at four meters from the discharge point, while the ship is stationary.

• All sampled pH values at discharge are above the calculated minimum pH

• Turbidity should not be greater than 25 FNU (formazin nephlometric units) or 25 NTU (nephlometric turbidity units) or equivalent units, above the incoming seawater turbidity.

• All wash water limits are continuous; these limitations must be upheld at any given time.

Nitrate wash-water limits

Flow rate from EGCS (t/MWh)Discharge Concentration

Limit (mg/L nitrate)

0 - 1 2700

2.5 1080

5 640

11.25 240

22.5 120

45 60

90 30

PAHs wash-water limits

Flow rate from EGCS (t/MWh)Discharge Concentration Limit (μg/L PAHphe equivalents)

0 - 1 2250

2.5 900

5 450

11.25 200

22.5 100

45 50

90 25

Strictest limits

*In this assessment, the strictest Nitrate and PAH limits have been selected for

comparison to all samples, rather than normalizing for operating conditions.

31

(Phenanthrene)

Total detected PAH compounds

0%

20%

40%

60%

80%

100%

PAH Nitrate

Average gross PAH and nitrate concentrations as % of IMO criteria

• The wash-water discharge

limitations vary according to the

EGCS wash-water output, i.e. the

higher the wash-water flow the

lower the allowed concentration.

• Here, the weighted average gross

post-EGCS concentrations are

compared with the lowest

theoretical requirements.

• Both the Post-EGCS PAH values for

phenanthrene and the average

sum of all detected PAH values

are below the strictest IMO

requirements.

• This comparison is for illustrative

purposes, and does not constitute

approval of any vessel with IMO

requirements.

CompoundLowest IMO criteria

(based on 90 t/MWh flow rate)Average gross post EGCS

concentration

PAH 25 µg/L 1.77 µg/L *

Nitrate 30 mg/L 0.8 mg/L

Note: IMO requirements for PAH are based on phenanthrene equivalence, so the post-EGCS value given is

for phenanthrene. Total detected PAH compounds are shown on the graph for information.

Comparison to IMO PAH and Nitrate criteria

Selection of Water Quality Standards

32

• As previously discussed, although not required for EGCS wash-water, a comparison to other water quality standards provides a clearer perspective and understanding of the parameters:

• German waste water ordinance (Article 2 of 6th Ordinance for Amendment of Waste Water Ordinance, Federal Water Act)

• EU incineration waste water standards (Directive 2010/75/EU, Industrial Emissions Directive)

• EU surface water standards (Directive 2013/39/EU, amending Water Framework Directive 2000/60/EU)

• WHO drinking water standards (Guidelines for Drinking-water quality, Fourth Edition)

• These water standards are not requirements to which EGCS wash-water discharge must comply and are chosen because they provide relatable criteria for a number of the parameters of interest in this assessment. The surface water and drinking water standards in particular provide real world criteria for PAH concentrations.

• They illustrate the general quality of the wash-water discharge and provide a broader perspective than comparison to IMO requirements, or simply a compilation of the average concentration alone can provide.

• These comparisons do not constitute the basis for determining if water is compliant to the selected standards. In particular, the surface water and drinking water standards are not intended for use with industrial processes and should be seen as for reference only.

• Parameter results shown are non-weighted average net post-EGCS concentrations.

33

German Wastewater Ordinance, net post-EGCS analysis Difference between the post-EGCS and Inlet averages

▪ This comparison uses the German wastewater ordinance (Article 2 of 6th Ordinance for Amendment of Waste Water Ordinance, Federal Water Act), wastewater limitations from biological waste treatment,.

▪ This ordinance was chosen as a point source discharge standard and due to the total metal criteria being well aligned with the tested parameters.

▪ The waste treatment criteria did not include vanadium limits, so this criteria limit is for steam generation (from the same ordinance), in order for vanadium to be included as a metal of interest.

▪ The standard contains no PAH criteria.

1% 1% 0%4% 2% 1%

0%

20%

40%

60%

80%

100%

Net Post-EGCS concentration as % of German Wastewater Ordinance

avg 3ơ/CR 100% criteria limit

0% 0% 1% 1% 0% 0%

7%

0% 1%

0%

20%

40%

60%

80%

100%

Arsenic Cadmium Chromium Copper Lead Mercury Nickel Thalllium Zinc

Annex 6 part 5 of 2010/75/EU Industrial Emissions Directive


34

German Wastewater Ordinance, net post-EGCS analysis Difference between the post-EGCS and Inlet averages

▪ The Waste gas cleaning waterstandards (part of EU IndustrialEmissions Directive 2010/75/EU)refer to emission limit valuesapplied to point source dischargefrom waste water from thecleaning of waste gases fromincineration or co-incinerationplants.

▪ The standard includes limits for parameters similar to the German wastewater ordinance (with the exception of Vanadium).

▪ The standard contains no PAH criteria.

EU Water Framework Directive, net post-EGCS analysis Difference between the post-EGCS and Inlet averages

35

▪ The EU surface water standards (part of EU Water Directive 2013/39/EU) refer to maximum allowed concentration* in inland surface waters.

▪ Surface water criteria imply that the concentrations shall not reach the maximum concentrations. This comparison does not account for any potential accumulation effect on ambient water.

▪ While as a water quality standard it is not intended for application to point source discharges such as EGCS washwater, it provides an important point of reference for PAH concentrations

▪ The metal concentration criteria are for dissolved (D) metals.

87%

1%

80%

49%

55%

2%

72%

0%

20%

40%

60%

80%

100%

Net Post-EGCS concentration as % of EU Surface Water Standard


* Maximum allowable concentration is the maximum concentration a pollutant is recommended to have at any given time in the water body in question.

36

Positive preliminary results for wash-water filtration

While not discussed further in this assessment, the results for 11 samples with discharge filtration were alsoreviewed. Initial tests and laboratory analysis of filtered material and post filtration wash-water indicate aneffective reduction potential for the following parameters:

Parameter Effective filtration



Arsenic Anthracene Aluminium

Cadmium Benzo(a)anthrancene Boron

Chromium Benzo(b)fluoranthene Cobalt

Copper Benzo(g,h,i)perylene Manganese

Iron Chrysene Selenium

Lead Fluoranthene Thallium

Mercury Fluorene Total Organic Carbon

Nickel Napthalene

Vanadium Phenanthrene

Zinc Pyrene

Further tests with filtration are being conducted to quantify the effectiveness of filtration. This includesoperational trials to determine an optimum balance between filtration and EGCS effectiveness.

• Summary of results

• Handling of statistical outliers

• Conclusions

• Limitations

37


• Background




• Conclusions

Summary of Results of the Assessment

38

• As for the 2016 assessment using the smaller sample set, a significant number of sample parameters are “non-detects” below laboratory detection limits.

• The average net Post-EGCS for a number metal parameters is extremely low, with post-EGCS values close to inlet values: Arsenic, Cadmium, Lead, Mercury, Selenium, Thallium

• A number of PAH parameters also give a low net result: Benzo(a)pyrene, Benzo(g,h,i)perylene, Benzo(k)fluoranthene.*

• As with the 2016 study, the 2018 samples are below the criteria for the German Wastewater Ordinance on all tested parameters. A new comparison to the stricter EU incinerator waste water standard similarly shows parameter concentrations well below reference point source emission limits.

• When comparing to the EU Surface Water Standard, the 2018 sample set is below the criteria when a three-sigma deviation and trimmed mean statistical methods is used to exclude statistical outliers. When outliers are not excluded, an unweighted average of sample parameters exceeds the EU surface water criteria for Anthracene and Dissolved Nickel.

• While the current sample size (11) is too small to reach definitive conclusions, Wash-water quality appears further improved by enhanced system filtration.

* The distribution of PAH compounds in the water column is recommended as a subject of further investigation to determine whether sub-surface seawater inlet samples provide accurate background PAH levels.

39

• In the 2016 assessment, the Benzo(b)fluoranthene parameter exceeded the EU surface water standard criteria. At the time it was hypothesised that this was due to a single outlier result.

• For the 2018 assessment, results outside a three-sigma deviation from the mean are excluded. This affects for only a small proportion of samples and a number of these statistical outliers are present in the inlet as well as post-EGCS samples.

• When comparing to the EU Surface Water Standard, the expanded 2018 sample set is below the criteria when statistical outliers are excluded. When outliers are not excluded, an unweightedaverage of sample parameters exceeds the EU surface water criteria for Anthracene and Dissolved Nickel.

• It is not possible to reach definitive conclusions on the cause of outliers. However, typical contributions to occurrence of outliers may include:

o Laboratory: Gross error, wrong analytical method, or insufficiently low detection limits.

o Sample contamination/procedure: Error in sampling procedures introducing contamination.

o Handling errors: Exceeding hold times, not maintaining temperature, breach of custody.

Handling of Statistical Outliers

40

• Studied wash-water parameter concentrations are below the emission limits for land-based industrial point source waste water standards. This is not evidence of compliance with these standards (which are intended for a different regime and include other controls and limits for compliance than quality criteria). The comparison does however establish a point of reference to understand the quality of EGCS wash-water relative to other industrial discharges.

• Very low and negative net post-EGCS values for certain wash-water parameters indicate a minimal contribution to the concentration of these parameters from the EGCS process (Arsenic, Cadmium, Lead, Mercury, Selenium, Thallium).

• The wash-water parameter concentrations also compare favourably to stricter water quality standards such as the EU surface water standards for inland waters. These standards have very different application and this comparison is not a definitive assessment of point source wash-water quality, however they provide a useful quantitative reference to understand wash-water parameter concentrations, in particular for PAH compounds where there is a lack of more suitable standards.

Conclusions

41

• The average net Post-EGCS concentrations were calculated for ease of comparison, but this means that values may not be representative of an individual vessel, given the variance of the samples, and various operational conditions of each vessel.

• The study provides an objective evaluation of wash-water quality as the basis for more informed debate. It does not:

• Attempt to assess the cumulative effect of wash-water parameters entering seawater or the potential environmental impact.

• Make any conclusions regarding wash-water parameter concentrations and fuel types, engine and flow conditions.

Limitations

Appendices

• Parameter value summary table

• Water standards and application in this study

• Average net Post-EGCS concentrations compared to German wastewater ordinance

• Average net Post-EGCS concentrations compared to EU incineration waste water standard

• Average net Post-EGCS concentrations compared to EU surface water standard

42

Inlet Post-DeSOx Average net

avg avg 3ơ # samples excluded

%samples excluded

avg avg 3ơ # samples excluded

%samples excluded

avg avg 3ơ

Acenaphthene 0.18171 0.12758 3 1.1% 0.25662 0.20525 3 1.1% 0.07491 0.07767

Acenaphthylene 0.18048 0.12633 3 1.1% 0.20132 0.14228 4 1.4% 0.02083 0.01595

Anthracene 0.27801 0.13083 2 0.7% 0.44387 0.21763 3 1.1% 0.16586 0.08680

Benzo(a)anthracene 0.12208 0.06727 3 1.1% 0.23149 0.10764 3 1.1% 0.10941 0.04037

Benzo(a)pyrene 0.11746 0.06260 3 1.1% 0.11923 0.06637 3 1.1% 0.00177 0.00377

Benzo(b)fluoranthene 0.12144 0.06663 3 1.1% 0.13298 0.08027 3 1.1% 0.01154 0.01364

Benzo(g,h,i)perylene 0.12786 0.07270 3 1.1% 0.12986 0.07673 3 1.1% 0.00200 0.00403

Benzo(k)fluoranthene 0.12123 0.06641 3 1.1% 0.11794 0.06506 3 1.1% -0.00329 -0.00135

Chrysene 0.12316 0.06836 3 1.1% 0.29627 0.24533 3 1.1% 0.17312 0.17697

Dibenz(a,h)anthracene 0.13486 0.08020 3 1.1% 0.12891 0.07615 3 1.1% -0.00595 -0.00404

Fluoranthene 0.16575 0.11143 3 1.1% 0.25066 0.17756 3 1.1% 0.08491 0.06613

Fluorene 0.16641 0.11210 3 1.1% 0.59659 0.52363 3 1.1% 0.43018 0.41153

Indeno(1,2,3-cd)pyrene 0.13775 0.08312 3 1.1% 0.24845 0.13643 1 0.4% 0.11070 0.05331

Naphthalene 0.19390 0.13990 3 1.1% 3.33595 2.42469 2 0.7% 3.14205 2.28480

Phenanthrene 0.17797 0.12379 3 1.1% 2.03806 1.89087 6 2.1% 1.86009 1.76708

Pyrene 0.19935 0.10795 3 1.1% 0.35858 0.27221 3 1.1% 0.15924 0.16426

Arsenic 0.02507 0.01969 3 1.1% 0.02476 0.01960 3 1.1% -0.00030 -0.00009

Cadmium 0.01060 0.00510 3 1.1% 0.01036 0.00508 3 1.1% -0.00024 -0.00002

Chromium 0.01377 0.01297 3 1.1% 0.01960 0.01604 3 1.1% 0.00583 0.00307

Copper 0.10160 0.04600 3 1.1% 0.16653 0.04869 3 1.1% 0.06494 0.00269

Lead 0.01801 0.01259 3 1.1% 0.01713 0.01192 3 1.1% -0.00087 -0.00067

Mercury 0.00403 0.00014 2 0.7% 0.00201 0.00014 1 0.4% -0.00202 0.00001

Nickel 0.05195 0.03460 6 2.1% 0.10495 0.07032 2 0.7% 0.05300 0.03572

Selenium 0.04630 0.04120 3 1.1% 0.04564 0.04073 3 1.1% -0.00066 -0.00047

Thallium 0.02492 0.01953 3 1.1% 0.02313 0.01793 3 1.1% -0.00179 -0.00159

Vanadium 0.05265 0.04762 3 1.1% 0.14099 0.12623 2 0.7% 0.08834 0.07860

Zinc 0.05020 0.03447 9 3.2% 0.14000 0.05479 1 0.4% 0.08981 0.02032

Arsenic Dissolved 0.02452 0.01918 3 1.1% 0.02383 0.01869 3 1.1% -0.00069 -0.00048

Cadmium Dissolved 0.01122 0.00572 3 1.1% 0.01087 0.00560 3 1.1% -0.00034 -0.00013

Chromium Dissolved 0.01856 0.01315 3 1.1% 0.01942 0.01597 2 0.7% 0.00086 0.00282

Copper Dissolved 0.04424 0.03190 9 3.2% 0.29756 0.08362 1 0.4% 0.25332 0.05172

Lead Dissolved 0.01874 0.01334 3 1.1% 0.02005 0.01233 4 1.4% 0.00131 -0.00101

Mercury Dissolved 0.00407 0.00015 2 0.7% 0.00389 0.00014 2 0.7% -0.00017 -0.00001

Nickel Dissolved 0.04164 0.03649 3 1.1% 0.09363 0.06111 2 0.7% 0.05199 0.02462

Selenium Dissolved 0.04422 0.03910 3 1.1% 0.04688 0.03871 4 1.4% 0.00266 -0.00038

Thallium Dissolved 0.02641 0.01921 4 1.4% 0.02468 0.01949 3 1.1% -0.00174 0.00028

Vanadium Dissolved 0.05718 0.04863 4 1.4% 0.11724 0.10575 2 0.7% 0.06006 0.05712

Zinc Dissolved 0.04418 0.03543 5 1.8% 0.12217 0.05118 1 0.4% 0.07799 0.01575

Parameter Summary Table

Water standards and their application in this study

44

Standard name Standard reference Applicable waters Application in study

German waste water ordinance

Article 2 of 6th Ordinance for Amendment of Waste Water Ordinance, Federal Water Act

The ordinance covers waste water from different industrial processes. Criteria for waste water from biological waste treatment and steam generation are used in this study.

All parameters tested by SGS relevant to the standard are compared to biological waste water criteria.

Vanadium concentration is compared to criteria for waste water from steam generation.

EU waste gas cleaning water standards

Directive 2010/75/EUWaste water resulting from the cleaning of waste gases from waste incineration and waste co-incineration plants.

All parameters tested by SGS relevant to the standard are compared to these criteria

EU surface water standards

Directive 2013/39/EU

Surface waters in the EU are defined as inland water, except groundwater, transitional or coastal waters.

2013/39/EU defines the priority substances and limitation criteria for the Water Framework Directive

All parameters tested by SGS relevant to the standard are compared to these criteria. In particular the standards provide PAH criteria limits and stricter trace metal concnetrations.

WHO drinking water standards

Guidelines for Drinking-water quality (Fourth Edition)

Drinking water standards which are considered to be within the reach of all countries throughout the world

All parameters tested by SGS relevant to the standard are compared to these criteria. This provides stricter criteria than point source discharge concentration limits for trace metals.

The average net Post DeSOx concentrations compared to the German wastewater ordinance

45

Substance Criteria

Net Post-EGCS results

Unit

Avg. 3σ% of

criteria limitAvg. Avg. 3σ

Cadmium 0.1 -0.0002 -0.00002 mg/L 0 %

Chromium 0.5 0.0058 0.0031 mg/L 1 %

Copper 0.5 0.0649 0.0027 mg/L 1 %

Lead 0.5 -0.0009 -0.0007 mg/L 0 %

Zinc 2 0.0898 0.0203 mg/L 1 %

Arsenic 0.1 -0.0003 -0.0001 mg/L 0 %

Mercury 0.05 -0.0020 0.00001 mg/L 0 %

Nickel 1 0.0530 0.0357 mg/L 4 %

Vanadium 4 0.0883 0.0786 mg/L 2 %

• The table shows

concentrations of net Post-

EGCS discharge.

• The standards are taken

from the German

wastewater ordinance;

these are wastewater

limitations from biological

waste treatment.

• Ordinance is

administered by the

German Federal

ministry for the

Environment

• This ordinance was

chosen because it had

criteria limits for the

highest number of relevant

parameters for point

source discharge.

The average net Post DeSOx concentrations compared to EU Industrial Emissions Directive standards

46

Substance Criteria


Unit

Avg. 3σ% of

criteria limitAvg. Avg. 3σ

Cadmium 0.05 -0.0002 -0.00002 mg/L 0%

Chromium 0.50 0.0058 0.0031 mg/L 1%

Copper 0.50 0.0649 0.0027 mg/L 1%

Lead 0.20 -0.0009 -0.0007 mg/L 0%

Zinc 1.50 0.0898 0.0203 mg/L 1%

Arsenic 0.15 -0.0003 -0.0001 mg/L 0%

Mercury 0.03 -0.0020 0.00001 mg/L 0%

Nickel 0.50 0.0530 0.0357 mg/L 7%

Thallium 0.05 -0.0018 -0.0016 mg/L 0%

• The criteria are based on

Annex 6 part 5 of

2010/75/EU Industrial

Emissions Directive -

Emission limit values for

discharges of waste water

from the cleaning of waste

gases from incineration

plants.

The average net Post DeSOx concentrations compared to the EU surface water standards

47

Substance Criteria


UnitNet Post

DeSOx results as % of criteriaAvg. Avg. 3σ

Anthracene 0.1 0.1659 0.0868 µg/L 87 %

Benzo(a)pyrene 0.270 0.0018 0.0038 µg/L 1 %

Benzo(b)fluoranthene 0.017 0.0115 0.0136 µg/L 80 %

Benzo(g,h,i)-perylene 0.0082 0.0020 0.0040 µg/L 49 %

Benzo(k)fluoranthene 0.017 -0.0033 -0.0013 µg/L -8 %

Fluoranthene 0.120 0.0849 0.0661 µg/L 55 %

Naphthalene 130 3.1420 2.2848 µg/L 2 %

Cadmium (Dissolved) 0.00045 -0.0003 -0.0001 mg/L -29 %

Lead (Dissolved) 0.014 0.0013 -0.0010 mg/L -7 %

Mercury (Dissolved) 0.00007 -0.0002 0.0000 mg/L -16 %

Nickel (Dissolved) 0.034 0.0520 0.0246 mg/L 72 %

• The EU surface water

standards refer to maximum

allowable concentration in

surface waters of the EU.

• Objective of the EU water

directive is to ensure that the

water quality does not

worsen with time.

• The cadmium criteria is

lower than what is possible

to measure during standard

lab testing.

The average net Post DeSOx concentrations compared to the WHO drinking water guidelines

48

EU drinking water standards

Criteria


Unit

Net Post DeSOxresults as % of

criteriaAvg. Avg. 3σ

Benzo(a)pyrene 0.7 0.0018 0.0038 µg/L 1 %

Arsenic 0.01 -0.0003 -0.0001 mg/L -1 %

Cadmium 0.003 -0.0002 0.0000 mg/L -1 %

Chromium 0.05 0.0058 0.0031 mg/L 6 %

Copper 2 0.0649 0.0027 mg/L 0 %

Lead 0.01 -0.0009 -0.0007 mg/L -7 %

Mercury 0.006 -0.0020 0.0000 mg/L 0 %

Nickel 0.07 0.0530 0.0357 mg/L 51%

Selenium 0.04 -0.0007 -0.0005 mg/L -1 %

Nitrate 50 1.2827 0.9030 mg/L 2 %

• These are not national

primary drinking water

regulations and are

selected as a

representative

international baseline.

• Criteria are only shown

for parameters for which

Carnival tested.

WHO Drinking Water Guidelines, net post-EGCS analysis Difference between the post-EGCS and Inlet averages

49

▪ This comparison is for general interest only and is not considered a relevant standard for EGCS wash-water

▪ Value for Nitrate is based on detected values only.

0%

20%

40%

60%

80%

100%

Net Post-EGCS concentration as % of WHO Drinking Water Guideline


Compilation and Assessment of Lab Samples from EGCS … · • In 2016 Carnival Corporation...

Documents

Transcript of Compilation and Assessment of Lab Samples from EGCS … · • In 2016 Carnival Corporation...